The overall goal of this application is to delineate the rules of human cortical plasticity, assayed by MEG/EEG based imaging, associated with auditory perceptual learning. Deciphering the rules governing cortical plasticity and perceptual learning is important for the development of training protocols aimed at improving auditory perception, which is impaired in a large proportion of individuals with learning-disability, reading-impairment, aging or other cognitive impairments. The conceptual theoretical framework of the current proposal is based on the Reverse Hierarchy Theory of Perceptual Learning (RHT). This theory has been successful in unifying several behavioral, anatomical and physiological findings of visual perceptual learning. We have adapted RHT for auditory perceptual learning and derived a series of testable hypotheses for spatiotemporal cortical plasticity and for the characteristics of behavioral improvement. RHT asserts that learning is a top-down guided process, which begins at high-level areas of the auditory processing hierarchy and progresses backwards along this hierarchy to earlier input levels so as to increase signal representation at the appropriate level in the hierarchy that is required for successful performance. Here, we test several predictions from RHT in the context of three tasks that sample three fundamental processes along the auditory processing hierarchy. These tasks are modulation-rate discrimination (R) and frequency discrimination (F) of simple and complex tone-trains to assess temporal, spectral and convergent spectrotemporal information processing. The first specific aim is to examine the rate and specificity of learning induced by brief and by extensive training protocols. The second specific aim is to examine characteristics of learning-induced plasticity, assayed by MEG/EEG based imaging. The third specific aim is to examine the effects of stimulus variability and psychophysical task conditions on cortical plasticity and learning characteristics. Our approach is to combine psychoacoustic studies to assay auditory perceptual learning with novel spatiotemporal source localization methods from MEG/EEG data. By comparing characteristics of behavioral improvements with cortical plasticity we propose to unravel general rules of cortical spatiotemporal plasticity associated with auditory perceptual learning.